Robust hybrid algorithms for regularization and variable selection in QSAR studies
This study introduces a robust hybrid sparse learning approach for regularization and variable selection. This approach comprises two distinct steps. In the initial step, we segment the original dataset into separate training and test sets and standardize the training data using its mean and standa...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nigerian Society of Physical Sciences
2023-11-01
|
| Series: | Journal of Nigerian Society of Physical Sciences |
| Subjects: | |
| Online Access: | https://journal.nsps.org.ng/index.php/jnsps/article/view/1708 |
| _version_ | 1797317287364001792 |
|---|---|
| author | Christian N. Nwaeme Adewale F. Lukman |
| author_facet | Christian N. Nwaeme Adewale F. Lukman |
| author_sort | Christian N. Nwaeme |
| collection | DOAJ |
| description |
This study introduces a robust hybrid sparse learning approach for regularization and variable selection. This approach comprises two distinct steps. In the initial step, we segment the original dataset into separate training and test sets and standardize the training data using its mean and standard deviation. We then employ either the LASSO or sparse LTS algorithm to analyze the training set, facilitating the selection of variables with non-zero coefficients as essential features for the new dataset. Secondly, the new dataset is divided into training and test sets. The training set is further divided into k folds and evaluated using a combination of Random Forest, Ridge, Lasso, and Support Vector Regression machine learning algorithms. We introduce novel hybrid methods and juxtapose their performance against existing techniques. To validate the efficacy of our proposed methods, we conduct a comprehensive simulation study and apply them to a real-life QSAR analysis. The findings unequivocally demonstrate the superior performance of our proposed estimator, with particular distinction accorded to SLTS+LASSO. In summary, the twostep robust hybrid sparse learning approach offers an effective regularization and variable selection applicable to a wide spectrum of real-world problems.
|
| first_indexed | 2024-03-08T03:32:44Z |
| format | Article |
| id | doaj.art-ee2ad4efc07c49fd9ec721ba2ffbeb2b |
| institution | Directory Open Access Journal |
| issn | 2714-2817 2714-4704 |
| language | English |
| last_indexed | 2024-03-08T03:32:44Z |
| publishDate | 2023-11-01 |
| publisher | Nigerian Society of Physical Sciences |
| record_format | Article |
| series | Journal of Nigerian Society of Physical Sciences |
| spelling | doaj.art-ee2ad4efc07c49fd9ec721ba2ffbeb2b2024-02-10T16:27:49ZengNigerian Society of Physical SciencesJournal of Nigerian Society of Physical Sciences2714-28172714-47042023-11-015410.46481/jnsps.2023.1708Robust hybrid algorithms for regularization and variable selection in QSAR studiesChristian N. Nwaeme0Adewale F. Lukman1African Institute for Mathematical Sciences, Mbour-Thies and BP. 1418, SenegalUniversity of Medical Sciences, Ondo State, PMB 536, Nigeria. | University of North Dakota, Grand Forks, ND, USA This study introduces a robust hybrid sparse learning approach for regularization and variable selection. This approach comprises two distinct steps. In the initial step, we segment the original dataset into separate training and test sets and standardize the training data using its mean and standard deviation. We then employ either the LASSO or sparse LTS algorithm to analyze the training set, facilitating the selection of variables with non-zero coefficients as essential features for the new dataset. Secondly, the new dataset is divided into training and test sets. The training set is further divided into k folds and evaluated using a combination of Random Forest, Ridge, Lasso, and Support Vector Regression machine learning algorithms. We introduce novel hybrid methods and juxtapose their performance against existing techniques. To validate the efficacy of our proposed methods, we conduct a comprehensive simulation study and apply them to a real-life QSAR analysis. The findings unequivocally demonstrate the superior performance of our proposed estimator, with particular distinction accorded to SLTS+LASSO. In summary, the twostep robust hybrid sparse learning approach offers an effective regularization and variable selection applicable to a wide spectrum of real-world problems. https://journal.nsps.org.ng/index.php/jnsps/article/view/1708High dimensionQSARMulticollinearityOutliersSparse Least trimmed squaresRandom forest |
| spellingShingle | Christian N. Nwaeme Adewale F. Lukman Robust hybrid algorithms for regularization and variable selection in QSAR studies Journal of Nigerian Society of Physical Sciences High dimension QSAR Multicollinearity Outliers Sparse Least trimmed squares Random forest |
| title | Robust hybrid algorithms for regularization and variable selection in QSAR studies |
| title_full | Robust hybrid algorithms for regularization and variable selection in QSAR studies |
| title_fullStr | Robust hybrid algorithms for regularization and variable selection in QSAR studies |
| title_full_unstemmed | Robust hybrid algorithms for regularization and variable selection in QSAR studies |
| title_short | Robust hybrid algorithms for regularization and variable selection in QSAR studies |
| title_sort | robust hybrid algorithms for regularization and variable selection in qsar studies |
| topic | High dimension QSAR Multicollinearity Outliers Sparse Least trimmed squares Random forest |
| url | https://journal.nsps.org.ng/index.php/jnsps/article/view/1708 |
| work_keys_str_mv | AT christiannnwaeme robusthybridalgorithmsforregularizationandvariableselectioninqsarstudies AT adewaleflukman robusthybridalgorithmsforregularizationandvariableselectioninqsarstudies |